Page 8 - Read Online

Page 446 Aledo-Serrano et al. J Transl Genet Genom 2021;5:443-55 https://dx.doi.org/10.20517/jtgg.2021.40

associated with other neurological phenotypes that do not include epilepsy, such as familial hemiplegic
[18]
migraine . This scenario, characterized by a wide variability in epilepsy phenotypes as well as non-epileptic
[19]
neurological phenotypes, has emerged for most epilepsy-associated genes . It is likely that some, but not
all, of the phenotypic variability is explained by gene variant-specific functional effects. For example, SCN1A
variants associated with hemiplegic migraine typically show gain of function. Other genetic and
environmental modifiers are likely to account for much of the remaining phenotypic variability, but these
factors remain difficult to be fully established and characterized. This has been shown in different research
works with functional studies, with relevant translational implications in genetic-phenotypic correlations,
such as the case of some biallelic variants in SCN1B .
[20]

Identification of new genetic variants
Targeted analyses focused on the identification of epilepsy-related genes can be separated into two types: (1)
those that compare huge cohorts of people with epilepsy with healthy individuals and search for a higher
proportion of common genetic variants among people with epilepsy, i.e., genome-wide association studies
(GWAS) ; and (2) those that look for rare deleterious variants, which might give an explanation of
[21]
[22]
patient’s clinical presentation in order to be considered as causative . Milder phenotypes are more
convenient for GWAS analysis, while severe phenotypes (DEEs) are suitable for the second strategy. De
novo pathogenic variants, which in most cases produce reproductive difficulties and are thus very rarely
found in the general healthy population, are often the most robust candidate variants .
[23]

GWAS studies do not try to search for disease causes. This approach normally tries to elucidate the
common genetic variants that confer a higher predisposition to epilepsy. This screening approach has been
[24]
valuable in understanding the gene-related architecture of epilepsy . Among these studies, scarce variants
have obtained genome-wide significance, and those variants explain a tiny percentage of the general
phenotypic variation . Thus, these research strategies show low clinical relevance. Future progress in this
[25]
direction will need larger numbers and less heterogeneity. In recent years, “polygenic risk scoring” strategies
are being developed attempting to determine the genetic load of common variants that together generate
more frequent and milder phenotypes, such as adult focal epilepsies or idiopathic generalized epilepsies .
[26]

Studies using next generation sequencing techniques to identify rare deleterious variants in deeply
phenotyped patients with severe neurodevelopmental disorders with epilepsy provide more useful
opportunities for new epilepsy-related genes discoveries . These approaches analyzed the epilepsy-related
[27]
disorders from two different perspectives. The first perspective involves the analysis of wide samples of
patients with broad phenotypes, with or without associated epilepsy, exemplified by the Deciphering
Developmental Disorders (DDD) study . The second involves the analysis of samples of individuals who
[28]
[29]
concretely presented epilepsy, such as the Epi4K consortium study . It is important to distinguish between
these two approaches, since, in the first one, epilepsy represents one of the symptoms included in a global
neurodevelopmental disorder, while, in the second one, epilepsy is the main selection criterion and
additional symptoms are presented as co-morbidities.

The illustrative case of STXBP1
Deleterious de novo variants in STXBP1 were firstly published in four patients with early-onset epilepsy,
with Ohtahara-syndrome features . Therefore, STXBP1 was firstly labeled as an “Ohtahara syndrome
[30]
gene”. Subsequent publications revealed that STXBP1 variants were associated with many other epilepsy
syndromes, such as Dravet, West or Lennox-Gastaut syndrome . Currently, STXBP1 phenotypic spectrum
[31]
is considered even broader, as a “neurodevelopmental disease-related gene”, since 11 patients of the DDD
study (out of 4293 families reported) had de novo deleterious STXBP1 variants . From this cohort, three
[32]
[33]
patients had no history of seizures and two others presented epilepsy in late childhood . Altogether, the